Measures for Assessment of Functioning and Outcomes in

Longitudinal Research on Child Abuse

Volume 2: Middle Childhood

Wanda M. Hunter, MPH Christine E. Cox, PhD Sarah Teagle, DrPH

Renee M. Johnson, MPH Ravi Mathew, MS

Elizabeth D. Knight, MSW Rebecca T. Leeb, PhD Jamie B. Smith, MA

October 2002

LONGSCAN Coordinating Center University of North Carolina at Chapel Hill

LONGSCAN has been supported by grants from the National Center on Child Abuse and Neglect and the Children’s Bureau, Administration on Children, Youth and Families, U.S.

Department of Health and Human Services; The National Institute of Child Health and Human Development; and the U.S. Department of Education

1

Abstract

LONGSCAN is a consortium of five longitudinal studies of the antecedents and

consequences of child abuse and neglect with common measurement and data collection

procedures. The consortium came together and developed interview protocols for measuring

etiologic and outcome variables related to child maltreatment in response to a call from the

National Research Council for longitudinal research in child abuse and neglect. This manual, the

second of two volumes, describes measures used by the LONGSCAN consortium at major

interviews at Ages 6 and 8 and at annual contact interviews at Ages 5, 7, 9, 10, and 11. The first

volume described measures used by the LONGSCAN consortium at the Age 4 interview, and at

major interviews before Age 4 for the two studies who joined the consortium with children

younger than 4. Each entry includes a description of the measure, its origin, administration and

scoring information, norms and comparative data, descriptive statistics from the LONGSCAN

samples, data on reliability and validity, and a copy of the instrument itself (as used in

LONGSCAN), when permissible. The manual serves as an essential reference for users of

LONGSCAN data, as well as for all researchers who seek useful measures related to child and

family well-being.

Table of Contents

Introduction ................................................................................................................... 2

Measures

1. After School Activity and Supervision………………………………………… 28

2. Behavioral Intent Scale (BIA)............................................................................. 33

3. Brief Symptom Inventory (BSI) ......................................................................... 43

4. Caregiver Demographics .................................................................................... 55

5. Caregiver's Physical Health Assessment ............................................................ 89

6. Caregiver Substance Use .................................................................................... 98

7. Center for Epidemiologic Studies Depression Scale (CES-D) ........................... 106

8. Child Aggressive Behavior ................................................................................. 116

9. Child Behavior Checklist/4-18 ........................................................................... 132

10. Child Behavior and Development ...................................................................... 148

11. Child Demographics ........................................................................................... 155

12. Child's Exposure to Substances .....................................................................…. 163

13. Child Health Assessment .................................................................................... 168

14. Child's Life Events .............................................................................................. 178

15. Child Sexual Behavior Inventory ....................................................................... 193

16. Child's Social Network Chart ............................................................................. 206

17. Conflict Tactics Scales: Caregiver to child ........................................................ 213

18. Conflict Tactics Scales: Partner-to-Partner ........................................................ 225

19. Discipline Methods ............................................................................................ 242

20. Duke-UNC Functional Social Support Questionnaire (FSSQ) ......................... 257

21. Everyday Stressors Index ................................................................................… 264

22. Father Involvement with Child ........................................................................... 272

23. Household Composition: Household Information Form and Family Chart ....... 282

24. Interviewer Ratings ............................................................................................ 297

25. Inventory of Supportive Figures ........................................................................ 314

26. Loneliness and Social Dissatisfaction Scale ...................................................... 327

27. My Family and Friends ..................................................................................... 334

28. Neighborhood Risk Assessment ....................................................................... 362

29. Neighborhood Short Form ................................................................................ 372

30. Pictorial Scale of Perceived Competence and Social Acceptance

for Young Children ............................................................................... 382

31. Preschool Symptom Self-Report (PRESS) ....................................................... 394

32. Revised Children's Manifest Anxiety Aales (RCMAS) ................................... 403

33. School Information Form ................................................................................. 415

34. School Safety Questionnaire ............................................................................ 424

35. Self-Report Family Inventory .......................................................................... 433

36. Service Utilization ........................................................................................... 445

37. Socially Desirable Response Set ...................................................................... 476

38. Social Provisions Scale .................................................................................... 483

39. Supplemental Service Utilization .................................................................... 493

40. Teacher's Estimation of Child's Peer Status ..................................................... 502

41. Teacher's Report Form ..................................................................................... 512

42. Things I Have Seen and Heard ........................................................................ 532

43. Trauma Symptom Checklist for Children - Alternate Version .....................… 561

44. Vineland Screener . .......................................................................................... 577

45. Wechsler Preschool and Primary Scale of Intelligence - Revised (WPPSI-R) ...................................................... 605

46. Welfare Reform Measure ............................................................................... 611

Acknowledgements

The authors acknowledge the contributions of LONGSCAN investigators and study

coordinators. We especially wish to recognize the LONGSCAN Measures Committee which has

included: Maureen Black, Christine Cox, Diana English, Mark Everson, Marcia Herman-

Giddens, Wanda Hunter, Jon Hussey, Elizabeth Knight, Alan Litrownik, Betsy Lowman, David

Marshall, Mary Wood Schneider, and Raymond Starr.

2

Introduction

Background

In 1991, the National Academy of Sciences, at the request of the Administration for

Children, Youth, and Families, charged an expert panel with evaluating the strengths and

weaknesses of current research on child maltreatment and developing a research agenda

(National Research Council, 1993). The panel concluded that methodology and instrumentation

issues were some of the most significant barriers to the conduct of useful research in the area of

child abuse and neglect. Their recommendations called for attention to the development of

techniques to measure individual experiences of maltreatment (as opposed to reliance on data

from agency records), use of well-established instrumentation to measure etiologic or outcome

variables, and the development of new measures accompanied by clear descriptions of the

measures' development, administration and scoring protocols, and information on reliability and

validity whenever possible (See Recommendation 8-7, p. 317).

Others have noted the extent to which inadequate measures have led to meaningless or

misleading results in maltreatment research (Fantuzzo & Twentyman, 1986; Milner, 1991).

Fantuzzo and his colleagues have specifically noted the absence of ecological validity and

inappropriate measurement in much of the research on maltreatment. In fact, lack of attention to

reliability, validity, and cultural sensitivity in the selection and development of research

instruments is a common criticism of social science research, especially in studies focused on

disadvantaged, minority populations (Garcia Coll et al., 1996; McGuire & Earls, 1993).

Ecological validity and cultural sensitivity are enhanced by gathering information in the child’s

typical contextual settings and from multiple natural respondents. Testing of maltreating families

in unfamiliar settings (clinics, shelters, crisis centers, etc.) may distort the responses of both

children and their parents (Fantuzzo & Twentyman, 1986).

There has also been inadequate attention paid to the processes by which effects occur,

such as the influence of age, social class, and gender. Cultural and developmental validity for

ethnic minority populations are dependent on the measurement tools that are used (Fantuzzo,

Weiss, Atkins, Meyers, & Noone, 1998). Reviews suggest that available rating scales for

preschool children have poor psychometric properties, especially with respect to low-income or

3

ethnic minority children (Garcia Coll, et al., 1996), probably because scale developers have

frequently failed to include these groups in test development.

The Consortium of Longitudinal Studies in Child Abuse and Neglect (LONGSCAN) was

formed in 1990, bringing together five long-term studies of child maltreatment which use

common procedures and instrumentation. The consortium has sought to identify or develop

appropriate instrumentation for the measurement of etiologic and outcome variables related to

child maltreatment. Several preexisting instruments have been modified, expanded, or otherwise

adapted for LONGSCAN use in an effort to improve measurement. This volume describes the

measures used at the Age 6, Age 8, and Annual Contact interviews at Ages 5, 7, 9, 10, and 11.

Descriptive and psychometric data are presented and summarized for each measure.

LONGSCAN - An Overview

Each of five LONGSCAN study sites (located in the East [EA], Midwest [MW], South

[SO], Southwest [SW], and Northwest [NW]) is conducting a separate and unique research

project on the etiology and impact of child maltreatment. While each project can stand alone on

its own merits, the use of common assessment measures, similar data collection methods and

schedules, and pooled analyses make LONGSCAN a collaborative effort that is greater than the

sum of its parts. In addition to the specific focus of the individual sites, the coordinated

LONGSCAN design permits a comprehensive exploration of many critical issues in child abuse

and neglect on a combined sample of sufficient size for unprecedented statistical power and

flexibility. Built into the LONGSCAN design is the ability to replicate and extend findings

across a variety of ethnic, social and economic subgroups.

Each site is following a sample of children who were identified in the first years of life as

being maltreated or at high risk for maltreatment. The findings of LONGSCAN will provide a

scientific basis for policy-making, program planning, and targeting service delivery by

increasing our understanding of the following:

* the child, family, and community factors which increase the risk for maltreatment

in its different forms;

* the differential consequences of maltreatment, depending upon its timing,

duration, severity, and nature, and upon the child's age and cultural environment;

4

* the child, family, and community factors (e.g., chronic exposure to violence,

parental substance abuse) that increase the harm caused by different forms of maltreatment;

* the factors that increase the probability of positive child outcomes despite

maltreatment and other adverse life circumstances;

* the strengths and weaknesses of various societal interventions such as child

welfare programs, foster care, mental health services, parenting classes, etc. Some of the sites are

involved in intervention research and evaluation of services, expediting the integration of

research findings into policy and practice.

The goal of LONGSCAN is to follow these children and their families until the children

themselves become young adults. Comprehensive assessments of children, their parents, and

their teachers are scheduled to occur at recruitment into LONGSCAN (where ages of children

range from infancy to five years old) and again at child ages 4, 6, 8, 12, 14, 16, and in young

adulthood. The first interview following enrollment into the LONGSCAN protocol was

scheduled to occur within three months before or after the child's 4th birthday, except at two

sites, NW and MW, which administered a pre-Age 4 interview in addition to the Age 4

interview. Because of delays in start-up, some children, mostly at the SO site, were older than 4

years of age at the first interview. Also, because of attrition after the first initial interview, some

families were recruited for the study at the Age 6 interview.

Maltreatment data are collected from multiple sources, including Child Protective

Services (CPS) and state Central Registry records, at least every two years. Yearly telephone

interviews allow the sites to track families and assess yearly service utilization and important life

events. A more detailed description of the Consortium’s conceptual model, methods, and

organization can be found in Runyan et al. (1998).

The LONGSCAN samples

The five LONGSCAN samples vary in their risk status and their exposure to

maltreatment. At the most extreme level, the SW site focuses exclusively on children who were

removed from their families and placed into foster care during the first 42 months of life because

of child maltreatment (See Landsverk, Davis, Ganger, Newton, & Johnson, 1996). Children

enrolled in the LONGSCAN protocol are followed, regardless of whether they remain in the

foster care system or return to their family of origin. Even though data were collected on the SW

5

sample in an earlier study, these children were first enrolled into the LONGSCAN protocol at

Age 4.

Three of the sites recruited children based on referral to CPS. In the NW sample, all

LONGSCAN participants were reported to CPS when they were less than 5 years old and were

judged to be at moderate risk for maltreatment based on a state risk assessment system. The MW

site recruited three groups of infants, two of which were from families who had recently been

reported to CPS for maltreatment. Of these two groups, one received six months of therapeutic

intervention and the other received standard care. The third group, a matched comparison group,

consisted of neighborhood families, who were matched according to children's age, ethnicity,

and family socioeconomic status (see Curtis, Schneider, & Calica, 1995). All children in the SO

site were originally recruited from a statewide High Priority Infant Program, which defined risk

status based on medical and sociodemographic criteria (see Kotch et al., 1995; Kotch, Browne,

Ringwalt, Dufort, Stewart, & Jung, 1997). Like the SW study, the SO site had collected data on

these children prior to the onset of LONGSCAN. When the SO site joined the LONGSCAN

Consortium, two groups of children were selected from the original sample for long-term study.

One group included children who had been reported to CPS prior to their fourth birthdays. The

other, a 2:1 comparison group, consisted of non-reported children who were matched with the

reported children on income, race, and sex.

The EA site includes low-income children who were recruited during infancy from

primary health care clinics (see Black, Dubowitz, Hutcheson, Berenson-Howard, & Starr, 1995;

Black, Hutcheson, Dubowitz, & Berenson-Howard, 1994; Black, Hutcheson, Dubowitz, Starr, &

Berenson-Howard, 1996). At recruitment, there were two risk groups: one defined by a child

factor (inadequate growth in the first two years of life), and the other defined by a parent factor

(HIV infection or drug use). The comparison group had no identified risk factors beyond their

low-income status. This pre-existing sample joined the LONGSCAN Consortium when the

children were four years of age.

Sociodemographic characteristics of the children and their primary caregivers in the five

samples at the time of the initial LONGSCAN interview are displayed in Tables 1 and 2. Racial

differences across the five samples are notable. The proportion of Black children ranged from

21% at the NW site to 93% at the EA site. The proportion of White children ranged from 5% at

the EA site to 51% at the NW site. Only the MW and SW sites had significant percentages of

6

Hispanic children, 15% and 16% respectively. The percentages of children with mixed racial

backgrounds were higher than for their caregivers across all sites, especially for the MW, SW

and NW sites. The SO site had more children characterized by low birth weight (31%) than the

other four sites, probably because low birth weight was one of the risk factors that made a child

eligible for recruitment at that site.

Tables 1 and 2 about here

The primary caregivers (who served as principal respondents at the pre-Age 4 and Age 4

interviews) tended to be the children’s biological mothers at all sites but the SW site (only 31%

biological mothers), where children were recruited based on early placement into foster care. In

the NW sample, the percentage of biological mothers who were primary caregivers at the initial

LONGSCAN interview was only 73%, compared to 89% in the SO sample, 91% in the EA

sample, and 99% in the MW sample. The differences observed here are probably due to sample

characteristics at the NW and MW sites: all children at the NW site were recruited based on a

recent report to CPS, and all the children in the MW sample were infants at the time of the initial

interview. When biological mothers were not the primary caregivers, this role was most likely

filled by grandmothers, other female relatives, or fathers. Only the SW sample had a notable

percentage of foster mothers (19%). There was also wide variation in the educational and marital

status of the primary caregivers. The MW site had the highest proportion of caregivers who had

not finished high school (61%), and along with the EA site, the highest percentage of mothers

who had never been married (69%). In the SW sample, where there was the highest proportion of

substitute caregivers, educational status and marital status were higher than at other sites. More

than half the families at all the sites were receiving Medicaid assistance, and 47% (SW) to 80%

(EA) were receiving Aid to Families with Dependent Children (AFDC) reflecting the generally

low socioeconomic status of all the samples.

Application of ecological-developmental theory

The Consortium has incorporated the research recommendations of the National Research

Council (1993) by relying on an ecological-developmental framework to define the theoretical

domains, to determine the data collection schedule, and to construct the age-specific interview

protocols. Both ecological theory and extant empirical research have suggested salient risk and

7

protective factors to be examined at the child, parent, family, neighborhood, and cultural levels.

The longitudinal design of the project reflects the developmental changes in risk and protective

factors that occur as children grow and change from early childhood through young adult years.

Because factors and processes influencing resilience can be instrumental in intervention

programs, LONGSCAN investigators are committed to the investigation of outcomes suggesting

resilience as well as poorer outcomes. Figure 1 portrays LONGSCAN's conceptualization of the

ecological-developmental theory that guides data collection as the children experience changing

relationships with their social ecology over time.

LONGSCAN's Ecological-Developmental conceptual model about here

Age-specific data collection points were selected to correspond to critical periods in

children's development, regardless of their exposure to child maltreatment. Because the sample

includes children with prior maltreatment histories, children who will first experience

maltreatment during the course of the project, and children who may never experience

maltreatment, LONGSCAN has the opportunity to examine both antecedents and consequences.

Selection and development of LONGSCAN measures

In addition to choosing measures based on ecological-development theory, the

LONGSCAN investigators have also sought, whenever possible, to include only measures that

are easy to use, culturally sensitive, repeatable, reliable, and valid. Thus, measures selection has

been based on the following guidelines:

1. Measures of child characteristics and experiences at major data collection points

must reflect the salient developmental issues for children of that age (See Cicchetti & Toth,

1981, for conceptualization of the etiology and consequences of child abuse and neglect from a

developmental psychopathology perspective.).

2. Measurement batteries for each data collection point must reflect each layer of the

nested social systems within which children of that age typically grow and develop, with greater

emphasis on family variables in the early childhood years and increasing emphasis on

extrafamilial variables as the child ages.

3. Priority is given to well-established and standardized measures with documented

reliability and validity across settings, type of interview, and test population.

8

4. Multiple natural informants (typically parent, child, and teacher) are used,

whenever possible, to take into account possible situational specificity of child behavior and

performance in different settings and to assess constructs from differing perspectives. Data

gathered from different respondents on the same measure also allow for the examination of

instrument validity.

5. The research protocol includes multiple indicators of constructs and external

sources of information, such as geocodes to describe neighborhood economic, crime, and social

indices. Data on maltreatment are collected from official records as well as from the children and

caregivers themselves.

6. Whenever possible, measures are repeated over time for the domains that continue

to be theoretically important. Repeated measures allow for the detection of change in subjects or

in their environments from one data point to the next.

7. Instruments are selected, developed, and administered with attention to cultural

and ethnic relevance, and sensitivity and to other human subjects issues, such as safeguarding

privacy and avoiding potential trauma to subjects.

8. Whenever possible, interviews with children and their caregivers take place

within familiar environments (e.g. home or school) to enhance the likelihood of ecological

validity. Only the EA site has performed interviews within a clinic setting, but this clinic is

neighborhood-based and one to which the families have been coming for well-child care since

the birth of the subject children.

9. The protocol has been designed to incorporate assessments of social desirability

and ratings from the interviewers regarding the validity of the responses.

For the pre-Age 4 and Age 4 protocols, primary caregivers are the principal respondents.

At Age 6, primary caregivers continue to provide information on parent and household variables,

teachers complete forms on the children's academic performance and social adjustment, and the

children provide information related to their developmental status, mental health, self

perceptions, exposure to violence, and perceptions of support. At Age 8 the protocol incorporates

tasks of middle childhood, including relationships with family and peers, academic achievement,

and adaptive behavior (See table of measures administered from baseline through Age 14 at

http://www.sph.unc.edu/iprc/longscan/measures/measurestable.pdf). At Age 12, during the

transition to adolescence, youth are asked about personal experiences with maltreatment. This

9

age level was chosen partially because 12-year-old children have the cognitive and emotional

skills to make informed responses to health care issues (Weithorn & Campbell, 1982), and

should be able to answer sensitive questions about maltreatment. The Age 14 interview is funded

by a supplemental grant from NICHD, and focuses on lifetime experiences of parental neglect

and delinquent and other problem behaviors. At Age 16, the protocol will focus on educational

achievement, occupational plans, and relationships with peers. For the final interview in young

adulthood, measurement will be focused on the transition to adult roles including family,

financial, and community responsibility.

The cultural and developmental appropriateness of the questions is determined by

conducting pre-tests and qualitative interviews which assess the acceptability, sensitivity,

comprehensibility and comprehensiveness of the measures prior to adopting the final battery for

each major data collection point. The goal is to develop interview protocols that are culturally

and developmentally appropriate, but not so specific that they would be unique to only one

community. In keeping with the recommended emphasis on transactions between the child and

the social ecology, questions are focused on children's experiences within the family, school,

neighborhood, and service system. LONGSCAN also contributes toward the definition of

maltreatment by relying on an ecological-developmental framework to examine maltreatment

from children directly at age 12, in addition to gathering information on reports from others.

About This Manual

Purpose

This manual, the second of two volumes, provides descriptions of all measures used

during middle childhood, at major interviews at Ages 6 and 8, and Annual Contact interviews

through Age 11. Volume 1 described measures administered in early childhood, at the baseline

or Age 4 interviews. These manuals were prepared as essential references for LONGSCAN

investigators and other investigators who use LONGSCAN data. It should also serve as a useful

reference for all researchers who seek to assess variables related to child and family well-being,

especially those engaged in child maltreatment research.

10

Presentation

The measures included are presented alphabetically by measure name. Each entry

includes a description of the measure, as well as a copy of the instrument itself (as used in

LONGSCAN), when permissible. Copyrighted instruments are not included, but information is

provided within the description about how to obtain copies of these instruments and further

documentation and scoring information.

The instrument descriptions are organized in a standard format that includes in the title

heading the instrument’s name, authors, and development or publication dates. When the

measure was developed by LONGSCAN, the referenced author is LONGSCAN since the

development of any new measure was led by the LONGSCAN Measures Committee which

includes representatives from each site. The description of each measure includes the purpose,

conceptual organization, item origin or selection, materials needed, time required for

administration, and training needed for examiners. This overview is followed by scoring

information (i.e., scoring protocols as developed by author or in subsequent research studies

including score types and score interpretation); norms and comparative data (as reported by

author or other researchers), and psychometric support (as reported by author or other

researchers).

LONGSCAN Use

A separate section on LONGSCAN’s use of the measure follows the more general

description of its development and prior use in empirical studies. Here we describe specifically

how we have used each instrument, including data points at which the instrument is used; the

respondent or informant for the measure, information about different versions of the form that

were used at different data points and their corresponding mnemonics. The rationale for selecting

the measure is also given. A section on administration and scoring notes describes any variations

in administration or scoring from that specified by the authors or other researchers. We also

describe any deviations across LONGSCAN sites in the administration or scoring of the

instrument.

Descriptive statistics from LONGSCAN data are also provided. For comparison

purposes, data are presented by race and by study site. Pre-age 4 data are only presented for the

11

MW and NW sites because these were the only two sites that collected LONGSCAN data on

children younger than four years of age.

Evidence of instrument reliability and validity were examined using LONGSCAN data

and the results of these analyses are described. The degree of internal consistency of scales was

typically assessed by Cronbach’s alpha coefficient (Cronbach, 1951). Split-half reliability was

used when appropriate; for example, for developmental tests where items become increasingly

more difficult (e.g., Wechsler Preschool and Primary Scale of Intelligence—Revised).

When possible, we also report initial examinations of the validity of these measures using

data from the LONGSCAN samples. Criterion, concurrent, and predictive validity are reported

as correlation coefficients. In a few instances, construct validity is examined by using factor

analysis. Within LONGSCAN data, one can also look for evidence of internal validity when

comparing the association between risk and protective variables and maltreatment. The more

similar the groups are on variables other than maltreatment the greater the certainty that observed

differences in outcome are attributable to maltreatment.

Future instrumentation from LONGSCAN

The LONGSCAN ecological-developmental model suggests increasing emphasis on

child- or youth-self report as children age, concurrent with increasing emphasis on cultural,

community and peer risk and protective factors relative to family or primary caregiver variables.

In recognition of the sensitive nature of youth self-report of maltreatment or engagement in risk

behaviors adolescent measures are administered using a project-developed Audio-Computer

Assisted Self Interview (A-CASI) system from Age 12 forward. The Age 12, Age 14, and Age

16 interviews include extensive project-developed child self-report measures of maltreatment,

assault, and witnessed violence history. Other domains include the continued assessment of risk

and protective factors related to child outcomes, and assessments of key developmental tasks for

mid- to late adolescence. These include assessments of: developing autonomy and separation

from caregiver; the development of healthy self-identity; pro-social behavior vs. antisocial and

delinquent or violent behavior; peer relations; future orientation and preparation for adulthood,

including academic performance and school orientation; and aspiration and planning towards

vocation and other adult roles.

12

References

Black, M. M., Dubowitz, H., Hutcheson, J., Berenson-Howard, J., & Starr, R. H., Jr.

(1995). A randomized clinical trial of home intervention among children with failure to thrive.

Pediatrics, 95 (6), 807-814.

Black, M. M., Hutcheson, J. J., Dubowitz, H., & Berenson-Howard, J. (1994). Parenting

style and developmental status among children with non-organic failure to thrive. Journal of

Pediatric Psychology, 19 (6), 689-707.

Black, M. M., Hutcheson, J. J., Dubowitz, H., Starr, R. H., Jr., & Berenson-Howard, J.

(1996). The roots of competence: Mother-child interaction among low-income, urban, African-

American families. Applied Developmental Psychology, 17 (3), 367-391.

Cicchetti, D., & Toth S. L. (1995). A developmental psychopathology perspective on

child abuse and neglect. Journal of the American Academy of Child & Adolescent Psychiatry,

34, 541-565.

Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests.

Psychometrika, 16, 297-334.

Curtis, P. A., Schneider, M. W., & Calica, R. H. (1995). The Capella Project: Potential

impact on public child welfare policy and practices. Proceedings of the Eighth National Round

Table on Risk Assessment, San Francisco. Washington, DC: American Public Welfare

Association.

Fantuzzo, J & Twentyman, C. (1986). Child abuse and psychotherapy research: Merging

social concerns and empirical investigation. Professional Psychology: Research and Practice, 17,

375-380.

Fantuzzo, J. W., Weiss, A. D., Atkins, M., Meyers, R., & Noone, M. (1998). A

contextually relevant assessment of the impact of child maltreatment on the social competencies

of low-income urban children. Journal of the Academy of Child and Adolescent Psychiatry, 37,

1201-1208.

Garcia Coll, C., Lamberrty, G., Jenkins, R., McAdoo, H. P., Crnic, K., Wasik, B., &

Vazquez Garcia, H. (1996). An integrative model for the study of developmental competencies

in minority children. Child Development, 67, 1891-1914.

13

Kotch, J. B., Browne, D. C., Ringwalt, C. L., Dufort, V., Stewart, P. W., & Jung, J.-W.

(1997). Stress, social support, and substantiated maltreatment in the second and third years of

life. Child Abuse & Neglect, 21 (11), 1025-1037.

Kotch, J. B., Browne, D. C., Ringwalt, C. L., Stewart, P. W., Ruina, E., Holt, K.,

Lowman, B., & Jung, J.-W. (1995). Risk of child abuse or neglect in a cohort of low-income

children. Child Abuse & Neglect, 19 (9), 1115-1130.

Landsverk, J., Davis, L., Ganger, W., Newton, R., & Johnson, I. (1996). Impact of child

psychological functioning on reunification from out-of-home care. Children and Youth Services

Review, 18, 447-462.

McGuire, J. & Earls, F. (1993). Exploring the reliability of measures of family relations,

parental attitudes, and parent-child relations in a disadvantaged minority population. Journal of

Marriage and the Family, 55, 1042-1046.

Milner, J. S. (1991). Measuring parental personality characteristics and psychopathology

in child maltreatment research. In R. H. Starr & D. A. Wolfe (Eds.), The effects of child abuse

and neglect (pp. 164-185). New York: Guilford.

National Research Council (1993). Understanding child abuse and neglect. Washington,

DC: National Academy Press.

Runyan, D. K., Curtis, P. A. Hunter, W. M., Black, M. M., Kotch, J. B., Bangdiwala, S.,

Dubowitz, H. , English, D., Everson, M. D., & Landsverk, J. (1998). LONGSCAN: A

consortium for longitudinal studies of maltreatment and the lifecourse of children. Aggression

and Violent Behavior: A Review Journal, 3, 275-285.

Steckler, A. (1992). Integrating qualitative and quantitative methods: An introduction.

Health Education Quarterly, 19, 1-8.

Weithorn, L. A., & Campbell, S. B. (1982). The competency of children and adolescents

to make informed decisions. Child Development, 53, 1589-1598.

14

Table 1. Child Characteristics by Sample at LONGSCAN Baseline* (Does not include subjects added to samples at the Age 6 interview)

Characteristic EA

(232) MW (317)

SO (221)

SW (318)

NW (261)

Age 4 years old infants 5 years old 4 years old 0-4 years old

Birth Year 1987-1991 1990-1995 1986-1987 1987-1991 1988-1992

Sex

Male 54% 49% 45% 45% 51%

Female 46% 51% 55% 55% 49%

Race

Black 93% 49% 62% 37% 21%

White 5% 14% 37% 29% 52%

Hispanic <1% 15% -- 16% 2%

Multiracial 1% 20% 1% 15% 21%

Other 1% 2% -- 3% 4%

Low Birth Weight 19% 19% 31% 7% 14%

* Based on data collected through July 1996. The definition of the LONGSCAN sample has been changed since the time this table was generated due to the attrition of subjects before Age 4 and the addition of subjects at Age 6. The new sample sizes for each site are as follows: EA: 282, MW: 245, SO: 243, SD: 330, and NW: 254.

15

Table 2: Primary Caregiver Characteristics at LONGSCAN Baseline*

(Does not include subjects added to samples at the Age 6 interview)

Caregiver Characteristic

EA (232)

MW (317)

SO (221)

SW (318)

NW (261)

Relationship to Child Biological Mother 91% 99% 89% 31% 73% Adoptive Mother - - .5% 16% 1% Grandmother 4% - 6% 11% 7% Other Female/ Kin 2% .3% 2% 9% 4% Foster Mother 1% .3% 1% 19% 7% Other Female/ Non-kin .4% - - 6% .4%

Male Primary Caregiver 2% .6% 1% 8% 7% Race

Black 93% 53% 61% 36% 22% White 5% 23% 38% 36% 63% Hispanic .4% 15% .5% 17% 2% Mixed Race .4% 7% .5% 4% 8% Other 1% 3% .5% 7% 5%

Education < 11 years 44% 61% 43% 29% 39% 12 years 42% 26% 39% 30% 30% >12 years 14% 14% 18% 41% 31%

Marital Status Married 16% 19% 38% 50% 31% Single/ Never Married 69% 69% 45% 19% 38%

Separated/Divorced 13% 11% 17% 27% 31% Widowed 2% 1% .5% 4% .4%

Welfare Support AFDC 77% 80% 48% 47% 63% Medicaid 72% 80% 69% 63% 70%

* Based on data collected through July 1996. The definition of the LONGSCAN sample has been changed since the time this table was generated due to the attrition of subjects before Age 4 and the addition of subjects at Age 6. The new sample sizes for each site are as follows: EA: 282, MW: 245, SO: 243, SD: 330, and NW: 254.

16